Re: [PATCH v4 9/9] t3512/t3513: remove KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1

[Phillip Wood]

On 04/12/17 19:24, Stefan Beller wrote:
> On Fri, Nov 24, 2017 at 3:07 AM, Phillip Wood  
> wrote:
>> From: Phillip Wood 
>>
>> Now that the sequencer creates commits without forking 'git commit' it
>> does not see an empty commit in these tests which fixes the known
>> breakage. Note that logic for handling
>> KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1 is not removed from
>> lib-submodule-update.sh as it is still used by other tests.
>>
>> Signed-off-by: Phillip Wood 
>> ---
>>
>> Notes:
>> The output of the tests with after the previous patch looks like the
>> output of the merge tests (see below), so I'm hopeful that this is a
>> genuine fix, but someone who knows about submodules should check that
>> things are in fact working properly now.
> 
> Looking at the patch only (in combination with the history of the
> submodule tests,
> 283f56a40b (cherry-pick: add t3512 for submodule updates, 2014-06-15))
> this patch
> looks good to me. I wonder though if this needs to be squashed in another 
> commit
> to keep the test suite working for each patch and have no intermittent
> failure in the
> series.

Hi Stefan

Thanks for looking at this, it's good to know you think it's OK. I'll
leave it to Junio's discretion whether to squash this into the previous
patch

Thanks again

Phillip

> Thanks,
> Stefan
> 

Re: [PATCH v4 9/9] t3512/t3513: remove KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1

[Stefan Beller]

On Fri, Nov 24, 2017 at 3:07 AM, Phillip Wood  wrote:
> From: Phillip Wood 
>
> Now that the sequencer creates commits without forking 'git commit' it
> does not see an empty commit in these tests which fixes the known
> breakage. Note that logic for handling
> KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1 is not removed from
> lib-submodule-update.sh as it is still used by other tests.
>
> Signed-off-by: Phillip Wood 
> ---
>
> Notes:
> The output of the tests with after the previous patch looks like the
> output of the merge tests (see below), so I'm hopeful that this is a
> genuine fix, but someone who knows about submodules should check that
> things are in fact working properly now.

Looking at the patch only (in combination with the history of the
submodule tests,
283f56a40b (cherry-pick: add t3512 for submodule updates, 2014-06-15))
this patch
looks good to me. I wonder though if this needs to be squashed in another commit
to keep the test suite working for each patch and have no intermittent
failure in the
series.

Thanks,
Stefan

[PATCH v4 9/9] t3512/t3513: remove KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1

[Phillip Wood]

From: Phillip Wood 

Now that the sequencer creates commits without forking 'git commit' it
does not see an empty commit in these tests which fixes the known
breakage. Note that logic for handling
KNOWN_FAILURE_CHERRY_PICK_SEES_EMPTY_COMMIT=1 is not removed from
lib-submodule-update.sh as it is still used by other tests.

Signed-off-by: Phillip Wood 
---

Notes:
The output of the tests with after the previous patch looks like the
output of the merge tests (see below), so I'm hopeful that this is a
genuine fix, but someone who knows about submodules should check that
things are in fact working properly now.

t3512-cherry-pick-submodule.sh

expecting success:
prolog &&
reset_work_tree_to add_sub1 &&
(
cd submodule_update &&
git branch -t modify_sub1 origin/modify_sub1 &&
$command modify_sub1 &&
test_superproject_content origin/modify_sub1 &&
test_submodule_content sub1 origin/add_sub1 &&
git submodule update &&
test_submodule_content sub1 origin/modify_sub1
)

Cloning into 'submodule_update'...
done.
Switched to a new branch 'add_sub1'
Branch 'add_sub1' set up to track remote branch 'add_sub1' from 'origin'.
Submodule 'sub1' (/home/phil/Documents/src/git/t/trash 
directory.t3512-cherry-pick-submodule/submodule_update_sub1) registered for 
path 'sub1'
Cloning into '/home/phil/Documents/src/git/t/trash 
directory.t3512-cherry-pick-submodule/submodule_update/sub1'...
done.
Submodule path 'sub1': checked out 
'ce9efb76b6ff5beb56e70d3662695f3ecbd38330'
Branch 'modify_sub1' set up to track remote branch 'modify_sub1' from 
'origin'.
[add_sub1 e57a25c] Modify sub1
 Author: A U Thor 
 Date: Fri Nov 24 10:48:45 2017 +
Submodule path 'sub1': checked out 
'7c9cd6d138a7bb3145fc0c7fca1f403cbe89010e'
ok 11 - git cherry-pick: modified submodule does not update submodule work 
tree

expecting success:
prolog &&
reset_work_tree_to add_sub1 &&
(
cd submodule_update &&
git branch -t invalid_sub1 origin/invalid_sub1 &&
$command invalid_sub1 &&
test_superproject_content origin/invalid_sub1 &&
test_submodule_content sub1 origin/add_sub1 &&
test_must_fail git submodule update &&
test_submodule_content sub1 origin/add_sub1
)

Cloning into 'submodule_update'...
done.
Switched to a new branch 'add_sub1'
Branch 'add_sub1' set up to track remote branch 'add_sub1' from 'origin'.
Submodule 'sub1' (/home/phil/Documents/src/git/t/trash 
directory.t3512-cherry-pick-submodule/submodule_update_sub1) registered for 
path 'sub1'
Cloning into '/home/phil/Documents/src/git/t/trash 
directory.t3512-cherry-pick-submodule/submodule_update/sub1'...
done.
Submodule path 'sub1': checked out 
'ce9efb76b6ff5beb56e70d3662695f3ecbd38330'
Branch 'invalid_sub1' set up to track remote branch 'invalid_sub1' from 
'origin'.
[add_sub1 155695c] Invalid sub1 commit
 Author: A U Thor 
 Date: Fri Nov 24 10:48:45 2017 +
error: Server does not allow request for unadvertised object 
0123456789012345678901234567890123456789
Fetched in submodule path 'sub1', but it did not contain 
0123456789012345678901234567890123456789. Direct fetching of that commit failed.
ok 12 - git cherry-pick: modified submodule does not update submodule work 
tree to invalid commit

expecting success:
prolog &&
reset_work_tree_to invalid_sub1 &&
(
cd submodule_update &&
git branch -t valid_sub1 origin/valid_sub1 &&
$command valid_sub1 &&
test_superproject_content origin/valid_sub1 &&
test_dir_is_empty sub1 &&
git submodule update --init --recursive &&
test_submodule_content sub1 origin/valid_sub1
)

Cloning into 'submodule_update'...
done.
Switched to a new branch 'invalid_sub1'
Branch 'invalid_sub1' set up to track remote branch 'invalid_sub1' from 
'origin'.
fatal: Needed a single revision
Branch 'valid_sub1' set up to track remote branch 'valid_sub1' from 
'origin'.
[invalid_sub1 497299e] Revert "Invalid sub1 commit"
 Author: A U Thor 
 Date: Fri Nov 24 10:48:46 2017 +
Submodule 'sub1' (/home/phil/Documents/src/git/t/trash